INDEX

Explanations

verbs or phrases related to activation or being activated

oai_token-act-pair · gpt-3.5-turbo

instances of the word "activate" or its variations in different contexts

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

esan

-0.83

ago

-0.69

agged

-0.65

×Ĳ

-0.65

apest

-0.65

 Apost

-0.64

 Hague

-0.64

ographers

-0.63

 Abram

-0.63

elf

-0.63

POSITIVE LOGITS

 activation

0.95

 charcoal

0.89

activated

0.88

 activated

0.86

 activate

0.85

 activating

0.82

activation

0.81

 activates

0.79

activate

0.77

 lockdown

0.75

Activations Density 0.026%

verbs or phrases related to activation or being activated

instances of the word "activate" or its variations in different contexts

No Comments

No Known Activations